Employment and Earnings

2002 Ad

2002 Ad

Both payroll employment and the unemployment rate were little changed in August. Job gains in services, government, and construction were largely offset by losses in manufacturing and retail trade.


Both the unemployment rate, 5.7 percent, and the number of unemployed persons, 8.1 million, were little changed over the month. The jobless rates for the major worker groups–adult men (5.2 percent), adult women (4.9 percent), teenagers (17.2 percent), whites (5.1 percent), blacks (9.6 percent), and Hispanics (7.5 percent)–showed little or no change. (See tables A-3 and A-4.)

The number of persons unemployed 15 weeks or more was 2.8 million in August, down from the recent high of 3.1 million in June. (See table A-13.)

Total employment and the labor force

Total employment rose by 429,000 to 134.5 million in August, after seasonal adjustment. The employment-population ratio was up by 0.2 percentage point to 62.8 percent. The civilian labor force (142.6 million) and the labor force participation rate (66.6 percent) were essentially unchanged over the month. (See table A-3.)

About 6.8 million persons (not seasonally adjusted) held more than one job in August. These multiple jobholders represented 5.0 percent of the total employed. (See table A-37.)

Persons not in the labor force

About 1.4 million persons (not seasonally adjusted) were marginally attached to the labor force in August, essentially the same as a year earlier. These individuals reported that they wanted and were available for work and had looked for a job sometime in the prior 12 months. They were not counted as unemployed, however, because they had not actively searched for work in the 4 weeks preceding the survey. The number of discouraged workers was 372,000 in August. Discouraged workers, a subset of the marginally attached, were not currently looking for work specifically because they believed no jobs were available for them. (See table A-36.)

Industry payroll employment

Total nonfarm payroll employment was little changed (+39,000) in August at 130.8 million. Since its recent low in April, payroll employment has edged up by 162,000. (See table B-3.)

The services industry added 100,000 jobs in August. Employment in this industry has risen by 411,000 since February. Employment in health services rose by 26,000 in August, in line with the average monthly increase over the prior 12 months. The help supply industry, which provides workers to other businesses, added 51,000 jobs over the month, following a decline of 30,000 in July. Since its recent low point in February of this year, employment in help supply services has risen by 165,000.

Government employment rose by 41,000 over the month. The Federal Government added 20,000 jobs, mostly reflecting an increase in the number of Federal security personnel at airports. Employment in local government grew by 34,000 in August, due largely to a gain in local education. State education employment fell by 20,000, after increasing by the same amount in July.

Employment in construction increased by 34,000 in August. Despite this 1-month increase, the level of construction employment in August was essentially the same as in April.

Manufacturing employment declined by 68,000 in August; this compares with losses in the prior 4 months that averaged 18,000. In August, job losses were widespread, including substantial declines in electronic and other electrical equipment (-18,000) and industrial machinery and equipment (-13,000). After remaining fairly steady from January through July, employment in fabricated metal products decreased by 10,000 in August. Rubber and plastics manufacturing lost 7,000 jobs, offsetting the previous month’s increase.

Retail trade, which had shown little change on balance since February, lost 55,000 jobs in August. A decline in department store employment (-41,000) accounted for most of the drop.

Weekly hours

The average workweek for production or nonsupervisory workers on private nonfarm payrolls edged up by 0.1 hour in August to 34.1 hours, seasonally adjusted. This follows a decline of 0.3 hour in July. The manufacturing workweek also was up by 0.1 hour over the month to 40.8 hours. Manufacturing overtime rose by 0.2 hour to 4.2 hours. Both measures had declined in July. (See table B-8.)

The index of aggregate weekly hours of production or nonsupervisory workers on private nonfarm payrolls rose by 0.3 percent in August to 147.9 (1982 = 100). The manufacturing index was down by 0.2 percent over the month. (See table B-9.)

Hourly and weekly earnings

Average hourly earnings of production or nonsupervisory workers on private nonfarm payrolls increased by 4 cents in August to $14.82, seasonally adjusted. Average weekly earnings increased by 0.6 percent over the month to $505.36. Over the year, both average hourly earnings and average weekly earnings grew by 3.1 percent. (See table B-11.)

Upcoming Changes to Nonfarm Payroll Series

NAICS conversion. The nonfarm payroll series, produced from the Current Employment Statistics (CES) program, will be converted from the 1987 Standard Industrial Classification (SIC) basis to the 2002 North American Industry Classification System (NAICS) basis beginning March 2003 with the release of January 2003 State and metropolitan area estimates and beginning June 2003 with the release of May 2003 national estimates, The NAICS conversion involves major definitional changes to many of the currently published SIC-based series. After the conversion to NAICS, SIC-based series will no longer be produced or published. Historical time series will be reconstructed as part of the NAICS conversion process. At the national level, all published series will have a NAICS-based history extending back to at least January 1990. For total nonfarm and other high-level aggregates, NAICS history will begin in January 1939, the current beginning date for these series. For more detailed series, the starting date will vary depending on the scope of the definitional changes between SIC and NAICS. At the national level, the NAICS-based reconstruction effort will cover all CES published data types: All employees, women workers, production workers, average weekly hours, average hourly earnings, and derivative series (for example, indexes of aggregate weekly hours). At the State and metropolitan area level, however, the NAICS-based reconstruction effort will cover only the all employee series, which will have a NAICS-based history extending back to January 1990, except for total nonfarm employment series which will retain their current beginning dates. There will be no reconstruction of average weekly hours, average hourly earnings, or other data types; these series will begin in January 2001 on a NAICS basis.

Completion of the CES sample redesign. March and June 2003 also will mark the completion of the CES sample redesign phase-in. The redesign converts the CES from a quota-based sample to a probability-based sample. In March 2003, at the State and metropolitan area level, the transportation and public utilities; finance, insurance, and real estate; retail trade; and services industries will be converted to the new sample design. In June 2003, at the national level, the services industries will be converted to the new sample design; all other private sector industries have already been converted. The final stage of sample redesign phase-in may result in level shifts for average weekly hours, average hourly earnings, production worker, and women worker series. New levels for these series are being computed from the NAICS-based probability sample.

Concurrent seasonal adjustment. Also beginning in June 2003, at the national level, the CES program will convert to concurrent seasonal adjustment, which uses all available monthly estimates, including those for the current month, in developing seasonal factors. Currently, the CES program projects seasonal factors twice a year. With the introduction of concurrent seasonal adjustment, BLS will no longer publish seasonal factors for CES national estimates. These changes will be made only tit the national level; there will be no changes made to the seasonal adjustment procedures for State-level CES series.

Change to Federal Government series. Beginning in June 2003, the national CES series for Federal Government employment will be revised slightly in scope and definition due to a change in source data and estimation methods. The current national series is an end-of-month Federal employee count produced by the Office of Personnel Management, and it excludes some workers, mostly employees who work in Department of Defense-owned establishments such as military base commissaries. The CES national series will include these workers. Also, Federal Government employment will be estimated from a sample of Federal establishments, will be benchmarked annually to counts from unemployment insurance tax records, and will reflect employee counts as of the pay period including the 12th of the month, consistent with other CES industry series. The historical time series for Federal Government employment will be revised to reflect these changes. With these changes, data at the national level will now be estimated in the same way for the Federal Government as is currently done at the State and metropolitan area level.

Small domain models. With the full implementation of the CES sample redesign at the State and metropolitan area level in March 2003, some redistribution of sample from smaller to larger metropolitan areas will be needed in order to maximize the reliability of the statewide total nonfarm estimates. In order to sustain the viability of the employment series for smaller domains, primarily detailed industry series within the smallest metropolitan areas, the Bureau of Labor Statistics (BLS) has developed a small domain model, based on weighted least squares regression methodology. The model uses as input available sample, time series history, and additional information from the full State sample. In March 2003, the model will be implemented as the official estimator for the approximately 10 percent of CES published series that have insufficient sample for direct sample-based estimates.

Further information on upcoming changes to CES data series is available through the BLS public database on the Internet, via the CES homepages at (national estimates) or (State and metropolitan area estimates), or by calling 202-691-6555 (national estimates) or 202-691-6559 (State and metropolitan area estimates).

Chart 1. Nonfarm payroll employment, seasonally adjusted, 1998-2002


Chart 2. Unemployed rate, seasonally adjusted, 1998-2002